DOI 10.1140/epje/i2018-11644-5

Regular Article

Eur. Phys. J. E (2018) 41: 44 THE EUROPEANPHYSICAL JOURNAL E

The effect of gramicidin inclusions on the local order ofmembrane components�

Elise Azar1, Doru Constantin1,a, and Dror E. Warschawski2,3

1 Laboratoire de Physique des Solides, CNRS, Univ. Paris-Sud, Universite Paris-Saclay, 91405 Orsay Cedex, France2 UMR 7099, CNRS-Universite Paris Diderot, Institut de Biologie Physico-Chimique, Paris, France3 Departement de Chimie, Universite du Quebec a Montreal, P.O. Box 8888, Downtown Station, Montreal H3C 3P8, Canada

Received 17 July 2017 and Received in final form 20 February 2018Published online: 28 March 2018 – c© EDP Sciences / Societa Italiana di Fisica / Springer-Verlag 2018

Abstract. We study the local effect of the antimicrobial peptide Gramicidin A on bilayers composedof lipids or surfactants using nuclear magnetic resonance spectroscopy and wide-angle X-ray scattering,techniques that probe the orientational and positional order of the alkyl chains, respectively. The twotypes of order vary with temperature and peptide concentration in complex ways which depend on themembrane composition, highlighting the subtlety of the interaction between inclusions and the host bilayer.The amplitude of the variation is relatively low, indicating that the macroscopic constants used to describethe elasticity of the bilayer are unlikely to change with the addition of peptide.

1 Introduction

The effect on the cell membrane of inclusions (membraneproteins, antimicrobial peptides etc.) is a highly activefield of study in biophysics [1]. A very powerful principleemployed in describing the interaction between proteinsand membranes is that of hydrophobic matching [2, 3]. Itstates that proteins with a given hydrophobic length insertpreferentially into membranes with a similar hydrophobicthickness [4].

Many studies of the interaction used as inclusion theantimicrobial peptide (AMP) Gramicidin A (GramA),which is known [5,6] to deform (stretch or compress) hostmembranes to bring them closer to its own hydrophobiclength, so the hydrophobic matching mechanism is likelyrelevant. This perturbation of the membrane profile in-duces a repulsive interaction between the GramA poresin bilayers with various compositions [7] that can be ex-plained based on a complete elastic model [8].

This large-scale description raises however fundamen-tal questions about the “microscopic effect” of the inclu-sion, at the scale of the lipid or surfactant molecules com-posing the membrane. To what extent is their local ar-rangement perturbed by the inclusion? Is the continuouselastic model employed for bare membranes still valid?

In this paper, our goal is to investigate the influenceof GramA inclusions on the local order of the lipid orsurfactant chains. We combine two complementary tech-niques: wide-angle X-ray scattering (WAXS) gives access

� Supplementary material in the form of a .pdf file availablefrom the Journal web page athttps://doi.org/10.1140/epje/i2018-11644-5

a e-mail: doru.constantin@u-psud.fr

to the positional order between neighboring chains, whilenuclear magnetic resonance (NMR) is sensitive to the ori-entational order of chain segments, thus yielding a compre-hensive picture of the state of the membrane as a functionof the concentration of inclusions.

We study GramA inserted within bilayers com-posed of lipids with phosphocholine heads and satu-rated lipid chains: 1,2-dilauroyl-sn-glycero-3-phosphocho-line (DLPC) and 1,2-dimyristoyl-sn-glycero-3-phospho-choline (DMPC) or of single-chain surfactants with zwit-terionic or nonionic head groups: dodecyl dimethyl amineoxide (DDAO) and tetraethyleneglycol monododecyl ether(C12EO4), respectively, the hydrophobic length of DLPC(20.8 A) [5] DDAO (18.4 A) [6] and C12EO4 (18.8 A) [7]is shorter than that of GramA (22 A) [9], while DMPC(25.3 A) [5] is longer. Since all these molecules form bi-layers, and their hydrophobic length is close to thatof GramA, the latter is expected to adopt the na-tive helical dimer configuration described by Ketchem etal. [10], and not the intertwined double helices observedin methanol [11] or in SDS micelles [12].

As for many molecules containing hydrocarbon chains,the WAXS signal of lipid bilayers exhibits a distinctivepeak with position q0 ∼ 14 nm−1, indicative of the packingof these chains in the core of the membrane. Although afull description of the scattered intensity would require aninvolved model based on liquid state theory [13], the widthof the peak provides a quantitative measurement for thepositional order of the lipid chains: the longer the rangeof order, the narrower the peak.

The effect of peptide inclusions on the chain peak hasbeen studied for decades [14]. Systematic investigationshave shown that some AMPs (e.g., magainin) have a very

Page 2 of 9 Eur. Phys. J. E (2018) 41: 44

strong disrupting effect on the local order of the chains:the chain signal disappears almost completely for a mod-est concentration of inclusions [15–17]. With other pep-tides, the changes in peak position and width are moresubtle [18] and can even lead to a sharper chain peak (asfor the SARS coronavirus E protein [19]).

To our knowledge, however, no WAXS studies of theeffect of GramA on the chain signal have been published.

NMR can probe global and local order parameters invarious lipid phases and along the lipid chain. Deuterium(2H) NMR has been the method of choice since the 1970sand has proven very successful until today [20–23]. Theeffect of GramA on the order parameter of the lipid (orsurfactant) chains has already been studied by deuterium(2H) NMR in membranes composed of DMPC [21,24–26],DLPC [26] and DDAO [6], but not necessarily at the sametemperature, concentration or lipid position as studiedhere.

Here, we use a novel application of solid-state NMRunder magic-angle spinning (MAS) and dipolar recou-pling, called the Dipolar Recoupling On-Axis with Scalingand Shape Preservation (DROSS) [27]. It provides simi-lar information as 2H NMR, by recording simultaneouslythe isotropic 13C chemical shifts (at natural abundance)and the 13C-1H dipolar couplings at each carbon posi-tion along the lipid or surfactant chain and head groupregions. The (absolute value of the) 13C-1H orientationorder parameter SCH = 〈3 cos2 θ − 1〉/2, with θ the anglebetween the internuclear vector and the motional axis, isextracted from those dipolar couplings, and the variationof order profiles with temperature or cholesterol contenthas already been probed, with lipids that were difficult todeuterate [28, 29]. Using the same approach, we monitorthe lipid or surfactant order profile when membranes aredoped with different concentrations of gramicidin.

The main advantages of 13C over 2H are: the possibilityto study natural lipids, with no isotopic labeling, and thehigh spectral resolution provided by 13C-NMR, allowingthe observation of all carbons along the lipid in a single2D experiment. Segmental order parameters are deduced,via a simple equation, from the doublet splittings in thesecond dimension of the 2D spectra. The data treatmentis simple for nonspecialists and the sample preparation isvery easy since there is no need for isotopic enrichment. Allthese facts make this technique ideal to probe and studynew molecules and to be able to compare the results withthe ones obtained with other similar particles.

The downsides are the reduced precision in the mea-surement and the impossibility to extract data from lipidsin the gel phase. In particular, carbons at the interfacialregion of the lipids (at the glycerol backbone and at thetop of the acyl chains) are less sensitive to changes inmembrane rigidity, and while subtle changes can be de-tected with 2H-NMR, they are difficult to interpret with13C-NMR at these positions. Furthermore, the inefficiencyof the DROSS method in the gel phase would theoreticallyallow measuring the lipid order in fluid phases coexistingwith gel phases and quantifying the amount of lipids ineach phase. In our measurements, lipids in the gel phasewere not abundant enough to be detected.

2 Materials and methods

2.1 Sample preparation

The samples were prepared from stock solutions of lipid orsurfactant and, respectively, Gram A in isopropanol. Wemix the two solutions at the desired concentration andbriefly stir the vials using a tabletop vortexer. The result-ing solutions are then left to dry under vacuum at roomtemperature until all the solvent evaporates, as verifiedby repeated weighing. The absence of residual isopropanolwas cheked by 1H NMR.

We then add the desired amount of water and mixthe sample thoroughly using the vortexer and then bycentrifuging the vials back and forth. Phases containingDMPC and DLPC were prepared at full hydration (incontact with excess water). C12EO4 systems containedbetween 47 and 50 vol.% D2O (for NMR) and between46 and 47 vol.% H2O (for WAXS). DDAO systems con-tained between 17 and 18 vol.% D2O (for NMR) and be-tween 18 and 25 vol.% H2O (for WAXS with and withoutcholesterol). The GramA concentration is quantified bythe molar ratio P/L (peptide to lipid or surfactant) to beconsistent with the literature. Note, however, that a sameP/L corresponds to twice as many inclusions per chain inmembranes composed of single-chain surfactants than inlipid bilayers.

For WAXS, we used a microspatula to deposit smallamounts of sample in the opening of a glass X-ray cap-illary (WJM-Glas Muller GmbH, Berlin), 1.5 or 2mm indiameter and we centrifuged the capillary until the sam-ple moved to the bottom. We repeated the process un-til reaching a sample height of about 1.5 cm. The capil-lary was then either flame-sealed or closed using a gluegun. For NMR, approximately 100mg of GramA/lipid orGramA/surfactant dispersion in deuterated water were in-troduced in a 4mm diameter rotor for solid-state NMR.

2.2 NMR

NMR experiments with DMPC, DLPC and C12EO4 wereperformed with a Bruker AVANCE 400WB NMR spec-trometer (1H resonance at 400MHz, 13C resonance at100MHz) using a Bruker 4mm MAS probe. NMR ex-periments with DDAO were performed with a BrukerAVANCE 300WB NMR spectrometer (1H resonance at300MHz, 13C resonance at 75MHz) using a Bruker 4mmMAS probe. All experiments were performed at 30 ◦C.

The DROSS pulse sequence [27] with a scaling factorχ = 0.393 was used with carefully set pulse lengths and re-focused insensitive nuclei enhanced by polarization trans-fer (RINEPT) with delays set to 1/8J and 1/4J and a Jvalue of 125Hz. The spinning rate was set at 5 kHz, typi-cal pulse lengths were 13C (90◦) = 3μs, 1H (90◦) = 2.5μsand 1H two-pulse phase-modulation (TPPM) decouplingwas performed at 50 kHz with a phase-modulation angleof 15◦.

1D spectra were acquired using the simple 13C-RINEPT sequence with the same parameters. For the 2D

Eur. Phys. J. E (2018) 41: 44 Page 3 of 9

Fig. 1. Example of a 2D 1H-13C DROSS spectrum for GramA/C12EO4 with P/L = 0.118.

spectra, 64 free induction decays were acquired, with 64 to512 scans summed, a recycle delay of 3 s, a spectral widthof 32 kHz and 8000 complex points. The total acquisitiontime was between 2 and 14 h. The data were treated usingthe Bruker TopSpin 3.2 software.

Resonance assignments followed that of previouslypublished data [22, 24, 27, 30, 31], using the Cω−n conven-tion, where n is the total number of segments, decreasingfrom the terminal methyl segment, Cω, to the upper car-bonyl segment C1. This representation permits a segment-by-segment comparison of the chain regions. Backbone re-gions are assigned according to the stereospecific nomen-clature (sn) convention for the glycerol moiety. Phospho-choline head group carbons are given Greek (α, β, γ) letterdesignations. The internal reference was chosen to be theacyl chain terminal 13CH3 resonance assigned to 14 ppmfor all lipids and surfactants studied here.

Order parameters were extracted from the 2D DROSSspectra by measuring the dipolar splittings of the Pakedoublet at each carbon site. This splitting was convertedinto a dipolar coupling by taking the scaling factor χ intoaccount. The absolute value of the segmental order pa-rameter is an additional “scaling factor” χ′ of the staticdipolar coupling into the measured dipolar coupling. Sincethe static dipolar coupling, on the order of 20 kHz, is notknown with high precision for each carbon, we have ad-justed it empirically in the case of DMPC, by comparingit to previously determined values [22,27,30].

2.3 WAXS

We recorded the scattered intensity I as a function of thescattering vector q = 4π

λ sin(θ), where λ is the X-ray wave-length and 2θ is the angle between the incident and thescattered beams.

Lipids. X-ray scattering measurements on the GramA/DLPC and GramA/DMPC systems were performed at

Fig. 2. Dipolar coupling slices of the Cω−2 at 30 ◦C.

the ID02 beamline (ESRF, Grenoble), in a SAXS+WAXSconfiguration, at an X-ray energy of 12.4 keV (λ = 1 A).The WAXS range was from 5 to 53 nm−1. We recordedthe integrated intensity I(q) and subtracted the scatter-ing signal of an empty capillary, as well as that of a watersample (weighted by the water volume fraction in the lipidsamples). We used nine peptide-to-lipid molar ratios P/Lranging from 0 to 1/5 and three temperature points: 20,30 and 40 ◦C.

Page 4 of 9 Eur. Phys. J. E (2018) 41: 44

Fig. 3. Orientational order parameter |SCH| for DMPC (a), DLPC (b), C12EO4 (c) and DDAO (d) bilayers embedded withGramA pores for different P/L at 30 ◦C. Error bars are smaller than symbol size.

The chain peak was fitted with a Lorentzian function:

I(q) =I0

( q−q0γ )2 + 1

.

We are mainly interested in the parameter γ, the half-width at half-maximum (HWHM) of the peak.

Surfactants. The GramA/DDAO and GramA/C12EO4

systems were studied using an in-house setup using assource a molybdenum rotating anode [32]. The X-ray en-ergy is 17.4 keV (λ = 0.71 A) and the sample-to-detectordistance is 75 cm, yielding an accessible q-range of 0.3 to30 nm−1. We used five peptide-to-surfactant molar ratios(also denoted by P/L) ranging from 0 to 1/5.5 and eighttemperature points, from 0 to 60 ◦C.

The best fit for the peak was obtained using a Gaussianfunction:

I(q) = I0 exp[− (q − q0)2

2σ2

].

For coherence with the measurements on lipid systems, wepresent the results in terms of the HWHM γ =

√2 ln 2σ.

We emphasize that the difference in peak shape(Lorentzian vs. Gaussian) is intrinsic to the systems(double-chain lipids vs. single-chain surfactants) and notdue to the resolution of the experimental setups, which ismuch better than the typical HWHM values measured.

3 Results and discussion

3.1 NMR

We acquired twelve 2D spectra for various surfactants andGramA concentration. Figure 1 shows the 2D DROSS

NMR spectrum of C12EO4 with a molar GramA concen-tration P/L = 0.118.

For each 2D spectrum, slices were extracted at eachcarbon position and order parameters were deduced. Fig-ure 2 shows a set of such representative slices (at the po-sition Cω−2).

As already explained, carbons at the glycerol backboneand at the first two positions along the acyl chains werediscarded. Figure 3 shows the order profiles determinedfor each lipid and surfactant, with variable amounts ofGramA.

As shown in fig. 3, there is hardly any change forthe head group region (Cα, Cβ and Cγ), which is ex-pected, considering the high mobility of this region, exceptin DDAO (CH3, C2 and C3). In the aliphatic region, inDMPC (fig. 3(a)), the order parameter increases for a ratioof P/L = 0.06 and then decreases for the P/L = 0.115. InDLPC and C12EO4 mixtures (figs. 3(b) and (c)), the or-der parameter slightly increases when adding the peptidecompared to the pure lipids with no significant depen-dence on P/L, reaching almost the same values for bothP/L = 0.053 and P/L = 0.112. For DDAO (fig. 3(d))we observe a remarkable increase in the order parameterprofile with increasing P/L all along the molecule but es-pecially in the acyl chain region.

Overall, we conclude that the order profiles signifi-cantly increase along the acyl chains with the concentra-tion of gramicidin, except in the case of DMPC where theorder profile globally increases with the addition of P/L =0.05 of gramicidin and then decreases at P/L = 0.11. Thispeculiar effect was already qualitatively observed by Riceand Oldfield, at the ω position by 2H NMR [24], and byCornell and Keniry, measuring the carbonyl CSA by 13CNMR [33]. The increase is larger in DLPC than in DMPC,as already observed by De Planque by 2H NMR with

Eur. Phys. J. E (2018) 41: 44 Page 5 of 9

2.0

1.5

1.0

0.5

0.0

HW

HM

[nm

-1]

0.200.150.100.050.00

P/L [mol/mol]

DMPC

20 °C 30 °C 40 °C

2.0

1.5

1.0

0.5

0.0

HW

HM

[nm

-1]

0.200.150.100.050.00

P/L [mol/mol]

DLPC

20 °C 30 °C 40 °C

Fig. 4. Width of the chain peak for DLPC (left) and DMPC (right) bilayers as a function of the GramA doping at threetemperatures.

P/L = 0.03 gramicidin [26]. The increase is also significantin DDAO, as observed by Oradd et al. by 2H NMR [6].In the head group region, effects are generally smaller,within the error bar, except for DDAO where we showthat gramicidin has the same effect as on the acyl chains.

Consequently, we show that gramicidin generally rigid-ifies the acyl chains of DLPC, C12EO4 and DDAO, as wellas the head group region of DDAO. In the case of DMPC,gramicidin first rigidifies the acyl chains, but more pep-tides tend to return the membrane to its original fluidity.

3.2 WAXS

The chain peak has long been used as a marker for the or-dered or disordered state of the hydrocarbon chains withinthe bilayer [34]. For lipids, an important parameter isthe main transition (or “chain melting”) temperature, atwhich the chains go from a gel to a liquid crystalline (inshort, “liquid”) phase [35]. The main transition tempera-ture of pure DLPC is at about −1 ◦C [36–39] and that ofpure DMPC is between 23 ◦C and 24 ◦C [36–40].

For the lipids, in the liquid phase the peak widthincreases slightly with P/L for all temperatures (fig. 4). Inthe gel phase of DMPC at 20 ◦C (fig. 4 (right) and fig. 5)this disordering effect is very pronounced, in agreementwith the results of several different techniques, reviewedin ref. [41] (sect. V-A). The linear increase in HWHMwith P/L can be interpreted as a broadening (ratherthan a shift) of the transition. The liquid crystallinephase value of the HWHM is reached only at the highestinvestigated P/L, amounting to one GramA molecule per5 or six lipids.

For surfactants, which we only studied in the liquidcrystalline phase, changes to the chain peak are slight. InC12EO4 membranes, the peak position q0 decreases veryslightly with temperature (fig. 6), while the peak width isalmost unchanged by temperature or gramicidin content(fig. 7 (right)). As an example, we observe a small decreaseof q0 with the temperature at P/L = 0.073 (fig. 6 (left)), aswell as a very slight increase with P/L at 20 ◦C, as seen infig. 6 (right). If we take the overall WAXS peak positionshift as a function of temperature and for all inclusionsconcentration (data not shown) we have a small temper-ature dependence for each P/L. Comparing the value in

65

60

55

50

45

40

35

30

25

20

15

10

5

0

I [1

0-3m

m-1

]

3530252015105

q [nm-1

]

Gram/DMPC 20°CP/L 1/5.21 1/6.26 1/7.82 1/10.43 1/14.6 1/21 1/31 1/52 0 (pure DMPC)

Fig. 5. Chain peak for DMPC in bilayers doped with varyingamounts of GramA at 20 ◦C.

absence of inclusion, the peak position slightly shifts afteradding gramicidin at a P/L = 0.015 but remains almostthe same for the different gramicidin content, showing nosignificant influence of the inclusions on the C12EO4 mem-branes.

This conclusion is confirmed by the very modestchange in the HWHM values presented in fig. 7 (right). AtP/L = 0, the HWHM is very close to 2.6 nm−1 for all tem-peratures. As the gramicidin content increases, we observea small gap between the different temperatures: the widthstays constant or increases for the lower temperatures (upto about 40 ◦C) and decreases for the higher ones. Thisgap widens at high gramicidin content (P/L > 0.07).

In the case of DDAO, the influence of gramicidin con-tent is more notable than for C12EO4 and the behavior isricher, especially in the presence of cholesterol.

Without cholesterol, the DDAO WAXS peaks coincidefor the different temperatures at a given inclusion concen-tration (e.g., in fig. 8 (left) at P/L = 0.178) whereas theprofiles differ according to the gramicidin concentrationfor a given temperature (see fig. 8 (right)).

These observations differ in presence of cholesterolwhere for one concentration of gramicidin inclusions (e.g.,case of P/L = 0.082 in fig. 9 (left)) at different temper-atures, we observe two families in which the spectra arequasi identical: one group at low temperatures (0–30 ◦C)

Page 6 of 9 Eur. Phys. J. E (2018) 41: 44

Fig. 6. Scattered signal I(q) for C12EO4 bilayers, as a function of temperature for a sample with P/L = 0.073 (left) and as afunction of concentration at room temperature: T = 20 ◦C (right).

Fig. 7. HWHM as a function of the concentration P/L, for all measured temperatures. DDAO bilayers (left) and C12EO4

bilayers (right).

Fig. 8. Scattered signal I(q) for DDAO bilayers, as a function of temperature for the most concentrated sample, with P/L =0.178 (left) and for all concentrations at T = 40 ◦C (right).

and another distinct group at higher temperatures (40–60 ◦C). At 20 ◦C, the peaks for DDAO cholesterol tend tosuperpose for P/L > 0.028 (data not shown), whereas at50 ◦C (fig. 9 (right)) the peak profiles differ and vary withP/L.

For the DDAO system, the peak occurs at much lowerq0 with cholesterol than without: q0 = 12.77 nm−1 at20 ◦C, 12.62 nm−1 at 30 ◦C and 12.28 nm−1 at 50 ◦C.Thus, the cholesterol expands DDAO bilayers, in con-trast with the condensing effect observed in lipid mem-branes [42, 43]. More detailed molecular-scale studieswould be needed to understand this phenomenon.

Without cholesterol, the width of the main peak inDDAO membranes is little affected by a temperaturechange, at least between 0 ◦C and 60 ◦C. Without grami-cidin, we observe two distinct HWHM values: ∼ 2.38 nm−1

at the lower temperatures (between 0 ◦C and room tem-perature) and ∼ 2.5 nm−1 for higher temperatures (be-tween 30 ◦C and 60 ◦C), but this gap closes with the addi-tion of gramicidin, and at high P/L only an insignificantdifference of 0.05 nm−1 persists (fig. 7 (left)).

On the other hand, at a given temperature the HWHMdoes vary as a function of P/L. This change is sigmoidal,with an average HWHM of ∼ 2.4 nm−1 for P/L < 0.05and ∼ 2.7 nm−1 for P/L > 0.11. Thus, above this concen-tration, the gramicidin decreases slightly the positionalorder of the chains.

An opposite effect is observed in the presence of choles-terol (fig. 10), where at high temperature (40–60 ◦C) theHWHM drops with the P/L: for instance, from 2.37 nm−1

to 2.08 nm−1 at 60 ◦C. At low temperature (0–30 ◦C) thereis no systematic dependence on P/L.

Eur. Phys. J. E (2018) 41: 44 Page 7 of 9

25

20

15

10

5

0

I(q)

[ar

b. u

nits

]

20151050

q [nm-1

]

DDAO CholesterolP/L=0.082

0°C 10°C 20°C 30°C 40°C 50°C 60°C

25

20

15

10

5

0

I(q)

[ar

b. u

nits

]

2520151050

q [nm-1

]

P/L 0 0.028 0.042 0.067 0.082

DDAO Cholesterol 50°C

Fig. 9. Scattered signal I(q) for DDAO/Cholesterol bilayers, as a function of temperature for a sample with P/L = 0.082 (left)and as a function of concentration at T = 50 ◦C (right).

Fig. 10. HWHM as a function of the concentration P/L, for allmeasured temperatures in the GramA/DDAO+Cholesterol/H2O system.

Overall we can conclude that gramicidin addition hasan effect that differs according to the membrane compo-sition. The temperature has a significant influence onlyin the presence of cholesterol. In all surfactant systemsand over the temperature range from 0 to 60 ◦C, the peakis broad, indicating that the alkyl chains are in the liquidcrystalline state. There are, however, subtle differences be-tween the different compositions, as detailed below.

In C12EO4 membranes, the peak position q0 decreasesvery slightly with temperature, while the HWHM is al-most unchanged by temperature or gramicidin content.

For DDAO (without cholesterol), q0 also decreaseswith temperature at a given P/L, but increases with P/Lat fixed temperature. On adding gramicidin, the HWHMincreases slightly with a sigmoidal dependence on P/L.Thus, a high gramicidin concentration P/L ≥ 0.1 reducesthe positional order of the chains in DDAO bilayers.

The opposite behavior is measured in DDAO mem-branes with cholesterol. Adding gramicidin inclusions havetwo distinct behaviors depending on the temperature. Forlow temperatures (between 0 ◦C and 30 ◦C) we have asmall peptide concentration dependence and a clear tem-perature correlation, whereas at high temperatures (be-tween 40 ◦C and 60 ◦C) we have a strong decrease in theHWHM in presence of inclusions depending only with the

P/L content without any variation with the temperaturerise. Since at P/L = 0 the HWHM value is very closefor the different temperatures then we can conclude thatadding gramicidin to a membrane containing cholesterolhelps rigidify it.

3.3 Comparing the NMR and WAXS results

Although the orientational and positional order param-eters are distinct physical parameters, one would expectthem to be correlated (e.g., straighter molecules can bemore tightly packed, as in the gel phase with respect tothe fluid phase.) This tendency is indeed observed in ourmeasurements, with the exception of DDAO.

We measured by NMR that the orientational order pa-rameter for DMPC increases when adding P/L = 0.05 andslightly decreases at P/L = 0.1 (fig. 3(a)). This behaviorwas also measured by WAXS for the positional order pa-rameter at both P/L values (fig. 4 (right)). Similarly, wemeasured for DLPC acyl chains the same orientationaland positional order profiles where the order increases forP/L = 0.05 and remains the same when adding P/L = 0.1gramicidin (figs. 3(b) and 4 (left)).

As for the C12EO4 surfactant acyl chains, we founda modest raise in both the orientational and the posi-tional order parameters when adding the gramicidin pep-tide with no dependence on the P/L molar ratio (figs. 3(c)and 7 (right)).

In the case of DDAO we found that adding gramicidinsignificantly increases the orientational order (fig. 3(d))and decreases the positional order (fig. 7 (left)). Solid-state NMR also shows an abrupt change in the head groupregion when little GramA is added, followed by a moregradual ordering of the acyl chain when more GramA isadded. This may imply a particular geometrical reorgan-isation of DDAO around the GramA inclusion that couldbe tested with molecular models.

4 Conclusions

Using solid-state NMR and wide-angle X-ray scattering,we showed that inserting Gramicidin A in lipid and sur-factant bilayers modifies the local order of the constituentacyl chains depending on multiple factors. In particular,

Page 8 of 9 Eur. Phys. J. E (2018) 41: 44

we studied the influence of membrane composition andtemperature on the local order.

The behavior of this local order is quite rich, withsignificant differences between lipids, on the one hand,and single-tail surfactants, on the other, but also betweenDDAO and all the other systems.

We showed that adding gramicidin influences the ori-entational order of the acyl chains and we find a similarbehavior for the orientational order and the positional or-der, except in the particular case of DDAO.

In this system, GramA content seems to notably influ-ence the DDAO acyl chains by decreasing their positionalorder and increasing their orientational order. GramA alsoinfluences the orientational order of the head groups. Alsoin DDAO, we showed by WAXS that the temperature hasa significant influence on the positional order only in thepresence of cholesterol.

In the gel phase of DMPC, GramA addition leads to alinear decrease in positional order, saturating at the liquidphase value for a molar ratio P/L between 1/6 and 1/5. Inthe liquid phase, we measure relatively small modificationsin the local order in terms of position and orientation whenadding Gramicidin A, especially in the case of DMPC,DLPC and C12EO4. This is a very significant result, whichallows further elaboration of elastic models in the presenceof inclusions by using the same elastic constants obtainedfor bare membranes.

As seen above for DDAO, in some membranes the pres-ence of inclusions influences differently the positional andorientational order of the acyl chains. Consequently, com-bining both techniques (NMR and WAXS) on the samesystem is very useful in obtaining a full image of the lo-cal order. A more detailed analysis could be performed bycomparing our results with molecular dynamics simula-tions. The correlation between changes in the chain orderand larger-scale parameters of the bilayer (e.g., the elasticproperties) could be established by using dynamic tech-niques, such as neutron spin echo.

We thank the CMCP (UPMC, CNRS, College de France) forthe use of their Bruker AVANCE 300 WB NMR spectrometer.We acknowledge ESRF for the provision of beamtime (experi-ment SC-2876) and Jeremie Gummel for his support. This workwas supported by the ANR under contract MEMINT (2012-BS04-0023). We also acknowledge B. Abecassis and O. Tachefor their support with the WAXS experiment on the MOMACsetup at the LPS.

Author contribution statement

DC and DEW designed research. All authors performedexperiments. EA and DC analyzed the data. All the au-thors contributed to the interpretation of the results andthe writing of the manuscript.

References

1. A. Lee, Biochim. Biophys. Acta 1612, 1 (2003).2. O.G. Mouritsen, M. Bloom, Biophys. J. 46, 141 (1984).3. J. Killian, Biochim. Biophys. Acta 1376, 401 (1998).

4. M.Ø. Jensen, O.G. Mouritsen, Biochim. Biophys. Acta1666, 205 (2004).

5. T.A. Harroun, W.T. Heller, T.M. Weiss, L. Yang, H.W.Huang, Biophys. J. 76, 937 (1999).

6. G. Oradd, G. Lindblom, G. Arvidson, K. Gunnarsson, Bio-phys. J. 68, 547 (1995).

7. D. Constantin, Biochim. Biophys. Acta 1788, 1782 (2009).8. A.F. Bitbol, D. Constantin, J.B. Fournier, PloS One 7,

e48306 (2012).9. J.R. Elliott, D. Needham, J.P. Dilger, D.A. Haydon,

Biochim. Biophys. Acta 735, 95 (1983).10. R. Ketchem, K.C. Lee, S. Huo, T. Cross, J. Biomol. NMR

8, 1 (1996).11. B. Wallace, K. Ravikumar, Science 241, 182 (1988).12. L.E. Townsley, W.A. Tucker, S. Sham, J.F. Hinton, Bio-

chemistry 40, 11676 (2001).13. A. Spaar, T. Salditt, Biophys. J. 85, 1576 (2003).14. G.W. Brady, D.B. Fein, Biophys. J. 26, 43 (1979).15. C. Munster, T. Salditt, M. Vogel, R. Siebrecht, J. Peisl,

Europhys. Lett. 46, 486 (1999).16. C. Munster, A. Spaar, B. Bechinger, T. Salditt, Biochim.

Biophys. Acta 1562, 37 (2002).17. A. Spaar, C. Munster, T. Salditt, Biophys. J. 87, 396

(2004).18. P.E. Schneggenburger, A. Beerlink, B. Weinhausen, T.

Salditt, U. Diederichsen, Eur. Biophys. J. 40, 417 (2011).19. Z. Khattari, G. Brotons, M. Akkawi, E. Arbely, I. Arkin,

T. Salditt, Biophys. J. 90, 2038 (2006).20. J. Davis, K. Jeffrey, M. Bloom, M. Valic, T. Higgs, Chem.

Phys. Lett. 42, 390 (1976).21. E. Oldfield, R. Gilmore, M. Glaser, H.S. Gutowsky, J.C.

Hshung, S.Y. Kang, T.E. King, M. Meadows, D. Rice,Proc. Natl. Acad. Sci. U.S.A. 75, 4657 (1978).

22. J. Douliez, A. Leonard, E. Dufourc, Biophys. J. 68, 1727(1995).

23. X.L. Warnet, M. Laadhari, A.A. Arnold, I. Marcotte, D.E.Warschawski, Biochim. Biophys. Acta 1858, 146 (2016).

24. D. Rice, E. Oldfield, Biochemistry 18, 3272 (1979).25. M.R. Morrow, J.H. Davis, Biochemistry 27, 2024 (1988).26. M.R.R. de Planque, D.V. Greathouse, R.E. Koeppe, H.

Schafer, D. Marsh, J.A. Killian, Biochemistry 37, 9333(1998).

27. J.D. Gross, D.E. Warschawski, R.G. Griffin, J. Am. Chem.Soc. 119, 796 (1997).

28. D.E. Warschawski, P.F. Devaux, J. Magn. Reson. 177, 166(2005).

29. D.E. Warschawski, P.F. Devaux, Eur. Biophys. J. 34, 987(2005).

30. A. Leftin, T.R. Molugu, C. Job, K. Beyer, M.F. Brown,Biophys. J. 107, 2274 (2014).

31. G. Oradd, G. Lindblom, Biophys. J. 87, 980 (2004).32. O. Tache, S. Rouziere, P. Joly, M. Amara, B. Fleury, A.

Thill, P. Launois, O. Spalla, B. Abecassis, J. Appl. Crys-tallogr. 49, 1624 (2016).

33. B. Cornell, M. Keniry, Biochim. Biophys. Acta 732, 705(1983).

34. A. Tardieu, V. Luzzati, F. Reman, J. Mol. Biol. 75, 711(1973).

35. R. Koynova, M. Caffrey, Biochim. Biophys. Acta 1376, 91(1998).

36. P.R. Cullis, M.J. Hope, New Compr. Biochem. 20, 1(1991).

37. B.A. Lewis, D.M. Engelman, J. Mol. Biol. 166, 203 (1983).

Eur. Phys. J. E (2018) 41: 44 Page 9 of 9

38. G. Cevc, D. Marsh, Phospholipid Bilayers: Physical Prin-ciples and Models (Wiley, 1987).

39. N. Kucerka, M.P. Nieh, J. Katsaras, Biochim. Biophys.Acta 1808, 2761 (2011).

40. J.F. Nagle, D.A. Wilkinson, Biophys. J. 23, 159 (1978).

41. J.A. Killian, Biochim. Biophys. Acta 1113, 391 (1992).42. D. Marsh, I.C. Smith, Biochim. Biophys. Acta 298, 133

(1973).43. W.C. Hung, M.T. Lee, F.Y. Chen, H.W. Huang, Biophys.

J. 92, 3960 (2007).